In 2006, the scientists who monitor methane, a greenhouse gas about 30 times more potent than carbon dioxide, thought that concentrations of the gas, which had sharply risen in the 1980s, had plateaued.

"If you look at the entire record from the beginning to 2006, it looks like a chemical system that is approaching steady state," said Edward Dlugokencky, an atmospheric chemist at the National Oceanic and Atmospheric Administration's Earth System Research Laboratory who monitors global methane emissions.

A puzzling aspect to this flattening of the methane trend was that, starting around 2000, China and other Asian countries were experiencing rapid development.

Typically, this would have resulted in increased methane emissions, Dlugokencky said.

"Chinese production of coal, which has methane emissions associated with it, had been increasing by about 7 percent per year since 2000, yet we see that methane is really flat in that period," he said. "So if there are increased anthropogenic emissions from increased fossil fuel use particularly in China, there has to be something that compensates for them."

The researcher hypothesized that methane emissions from wetlands, which are the biggest source of methane worldwide, were lower during that period, balancing out any increase from Asian development.

A mystery takes wing
As Dlugokencky and others write in a perspective paper published yesterday in the journal Science, accounting for the reasons behind trends in methane emissions is still an area of scientific uncertainty.

Since 2007, they report, methane has been on the rebound, with atmospheric concentrations growing quickly.

"There are a lot of puzzles as to exactly why it's growing," said Euan Nisbet, an earth scientist at the University of London who was the paper's lead author.

"Part of this article was trying to figure out the causes of the more recent growth, and there seem to be quite a lot of causes. And there's still some very big science problems as to what actually is going on," Nisbet added.

Scientists point to a few main reasons for the growth in methane over the past seven years. First, wet periods in the southern tropics have led to wetlands growing and lasting longer. That's one big source of methane.

Also, in 2007, methane emissions from the Arctic increased, probably because higher temperatures led to wetlands there releasing more of the gas.

A third reason for the increase in recent years is the growth in fossil-fuel-related emissions, Nisbet said.

If you can't measure it, can you reduce it?
"We've been using quite a lot more coal, especially in China, and more natural gas," he said. Both of those sources leak methane.

In the United States, researchers measuring methane leaks from natural gas have often come up with higher leak estimates than those of U.S. EPA, which uses a "bottom up" approach of multiplying a calculated emissions rate by the number of a certain type of source (ClimateWire, Nov. 26, 2013).

"Doing a bottom-up estimate is quite difficult," Nisbet said. "Because you are trying to calculate gas leaks, landfill leaks, you count up all the cows and you multiply by how much you think they are emitting. So it's a tough job."

One way to improve global methane measurements, the researchers said, is to increase the number of stations where methane is measured. In particular, tropical regions are lacking in measurements, and the Arctic could also benefit from additional measurements.

"The tropics are a very important region because so much is happening there," Dlugokencky said.

If methane emissions were better understood, policymakers could potentially reduce greenhouse gas emissions without too much effort, Nisbet added.

"You don't have to completely rearrange the entire world economy to [reduce methane emissions]. You've got to stop some gas leaks, you've got to stop some grass fires or stop landfill emissions. So methane is a very important target," Nisbet said.

"But we have to know where it is coming from. If we want to bring down the sources, we have to know where the sources are. And for that we need more measurements."

Reprinted from Climatewire with permission from Environment & Energy Publishing, LLC., 202-628-6500